Identidfying living arrangements of individuals

USA
1

Hi,

I am working on the ACS 2023 microdata set. I want to identify the following categories for individuals:

• Living alone
• Living only with a spouse
• Living only with a spouse and children
• Living only with children
• Living with others (related or unrelated)

I am using Stata 17.

I identified individuals living alone with the following formula:

replace livingalone = 1 if hhtype == 2 | hhtype == 3 | hhtype == 4 | hhtype == 6 & famsize == 1

To identify individuals living only with their spouses/partners, I used the following formula:

replace livingwspouse = 1 if hhtype == 1 & nchild == 0 & famsize == 2

For individuals living only with children:

replace livingwchild = 1 if nsubfam == 0 & momloc == 00 & poploc == 00 & momloc2 == 00 & poploc2 == 00 & nsibs == 0 & sploc == 0 & gchouse == 1 & (hhtype == 2 | hhtype == 3) & (nchild == 1 | nchild == 2 | nchild >= 3 | nchild == 4 | nchild == 5 | nchild == 6 | nchild == 7 | nchild == 8 | nchild == 9)

For individuals living with children and a spouse:

replace livingwsc = 1 if hhtype == 1 & nsubfam == 0 & momloc == 00 & poploc == 00 & momloc2 == 00 & poploc2 == 00 & nsibs == 0 & gchouse == 1 & (nchild == 1 | nchild == 2 | nchild == 3 | nchild == 4 | nchild == 5 | nchild == 6 | nchild == 7 | nchild == 8 | nchild == 9)

For individuals living with others:

replace livingwothers = 1 if (inlist(hhtype, 2, 3, 5, 7)) & sploc == 00 & nchild == 0 & inlist(relate, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)

I want to identify with whom individuals live based on the categories I created above. My results are probably/strongly incorrect because I could not choose the right variables to match members of the same household. Could you please help me with this?

2

HHTYPE is created by IPUMS in order to mirror the methodology used by the Census Bureau in classifying household types in the 2000 PUMS Census sample. Its main use is for historical comparability; there are a number of caveats with using this variable with recent data such as the classification of households with unmarried partners and same-sex partners (see the comparability tab for HHTYPE). As an alternative, IPUMS provides a number of family interrelationship variables (e.g., FAMUNIT, FAMSIZE, NCHILD, SPLOC) that incorporate these types of partnerships as well other social relationships that you may find more helpful for your research question. I’ve edited your code below to share how these might be used to identify the groups you mention.

Individuals living alone:
These are persons who do not reside with any family members (i.e., those related by blood, marriage/cohabitation, or adoption). They may reside with other non-related individuals.
replace livingalone = 1 if famsize == 1

Individuals living only with their spouses/partners:
replace livingwspouse = 1 if sploc >0 & famsize == 2

Individuals living only with children:
This includes biological, step, and adopted children of any age:
replace livingwchild = 1 if nchild > 0 & famsize == nchild + 1

For individuals living with children and a spouse (married or cohabiting):
replace livingwsc = 1 if sploc > 0 & nchild > 0 & famsize == nchild + 2

For individuals living with others:
These are persons who reside with anyone else, whether a family member or a non-related household member, except for a spouse or child
replace livingwothers = 1 if NUMPREC > 1 & sploc == 0 & nchild == 0

1 Like
3

Thank you very much! I appreciate your help!

4

Hi again,

I have several issues regarding these categories again.

  1. I applied the formulas you provided, but for some individuals, both living alone and living with others are marked as “1.”

  2. Should the total percentage of these categories equal 100 percent? If they are supposed to sum to 100 percent, then it seems the codes you provided do not work correctly.

  3. Also, I think there might be individuals who do not fall into any of the categories I created. For example, if an individual lives with a sibling and own children, could they fall outside the living with others category since the code specifies nchild == 0?

Thank you very much.

5

I apologize that I missed your response to this. I will follow-up with a response early next week. Thanks for your patience.

6

Hi,

That’s totally fine. I look forward to your response. Thank you very much!

Best,
Damla

7

Hi Damla,

The code we provided was intended to demonstrate how you might use these variables to capture the concepts you described, but I assume it will need to be adapted to your specific research application.

In your original post, you specified an interest in creating a series of variables to identify individuals in a number of different living arrangements. It sounds like you are now considering creating a singular variable that reports an individual’s living arrangement in mutually exclusive categories. This would certainly address some of the issues you raised (e.g., persons in living arrangements that are not captured by the variables you outlined previously, potential overlap between living alone and living with others). Additionally, defining a single variable that captures mutually exclusive categories of living arrangements would ensure that your percentages sum to 100.

While I highlighted a number of the IPUMS family interrelationship variables in my previous response, I didn’t describe the variable NUMPREC in detail. This simply reports the number of person records in a household and may be a helpful resources as you seek to compare the likely interrelationships of individuals against the total size of the household.